Re: CBA 1 Comments on the BCBS consultative document: Regulatory treatment of accounting provisions interim approach and transitional arrangements

Transcription

1 Box 348, Commerce Court West 199 Bay Street, 30 th Floor Toronto, Ontario, Canada M5L 1G2 Darren Hannah Vice-President Finance, Risk & Prudential Policy Tel: (416) Ext. 236 January 13, 2017 Secretariat of the Basel Committee on Banking Supervision (BCBS) Bank for International Settlements CH-4002 Basel, Switzerland Re: CBA 1 Comments on the BCBS consultative document: Regulatory treatment of accounting provisions interim approach and transitional arrangements We thank you for the opportunity to provide comments on the BCBS s consultative document, Regulatory treatment of accounting provisions interim approach and transitional arrangements ( consultative document ). In this letter, we provide overall comments on the proposals with more detailed comments included in the attached appendix. We support the Committee s proposal to retain the current regulatory treatment of accounting provisions, for an interim period, under both the standardized and internal ratings-based approaches. We agree with maintaining the status quo and specifically with the Committee s suggestion to allow local jurisdictions to extend their existing approaches to categorizing provisions as general or specific provisions measured under the applicable expected credit loss (ECL) model. This is important given the limited time until the effective adoption date of IFRS 9 by member jurisdictions. We have ongoing communication with our regulator, OSFI, on IFRS 9 implementation since Canada s Domestic Systemically Important Banks are required to implement the standard starting November 1, Overall, we support a transitional arrangement for the impact of ECL accounting provisions on regulatory capital and recommend that Approach 1 in the consultative document be adopted for reasons of simplicity, ease of implementation, and predictability. This is a significant accounting change with regulatory implications so we would advocate for a transition period of at least five years. We also believe that a materiality threshold under Approach 1 suggesting a modest decrease in CET1 should be absorbed by banks prior to the transition amount being determined is not warranted given the complexity of modelling associated with the ECL models. While we appreciate the BCBS s argument that banks capital planning should have taken some amount of impact into consideration, we suggest that the BCBS similarly take into consideration that the 1 The Canadian Bankers Association works on behalf of 59 domestic banks, foreign bank subsidiaries and foreign bank branches operating in Canada and their 280,000 employees. The CBA advocates for effective public policies that contribute to a sound, successful banking system that benefits Canadians and Canada's economy. The Association also promotes financial literacy to help Canadians make informed financial decisions and works with banks and law enforcement to help protect customers against financial crime and promote fraud awareness.

2 IFRS 9 impact estimate is an ongoing process that requires further refinement, and to the extent the regulatory capital rules are not finalized, the full regulatory capital impact will not be fully known until closer to the implementation date. Furthermore, we believe that the disclosure requirements around the transitional arrangement should be kept simple and not be overly prescriptive to avoid the undue burden of disclosing multiple layers of transitional arrangements and various threshold-related impacts. Please refer to our detailed comments accompanying this letter. In conclusion, we would stress that an immediate regulatory solution is needed with the upcoming implementation date for IFRS 9. We therefore support the Committee s proposal to maintain the current regulatory treatment of accounting provisions for an interim period with local adoption of the regulatory definitions of general and specific provisions. We appreciate that more time is needed to complete more fulsome work on a long-term approach and, as such, we also believe that a transitional arrangement should be provided at a minimum. We thank you in advance for your consideration of our comments, and we would be pleased to discuss our submission at your convenience. Sincerely, cc: Karen Stothers, Senior Director, Accounting, Accounting Policy Division, OSFI Jihan Bhuiyan, Senior Accounting Advisor, Accounting Policy Division, OSFI Catherine Girouard, Director, Bank Capital, Capital Division, OSFI 2

3 Overall Comments and Key Issues We appreciate the BCBS consideration of the magnitude of the IFRS 9 impact on CET1 and the need for a transitional arrangement. IFRS 9 represents a fundamental change to the accounting provision calculation and could result in a significant decrease in banks CET1 capital. Furthermore, there are also other regulatory reforms that are planned in the same time frame, which would further increase the pressure on banks regulatory capital ratios. As such, a transitional arrangement would help banks manage their capital ratios while planning for the implementation of all new BCBS standards. We believe that the phase-in period should be 5 years as the implementation of expected credit loss (ECL) models is a significant accounting change with regulatory implications. The reference metric should continue to be CET1 given the shortfall in allowance is currently reflected in CET1. Simplicity in application would be desired and would reduce any IT costs associated with implementation. No comments 1. Introduction (p. 1-2) 2.3 Proposal to retain the current regulatory treatment of provisions (p. 3-4) We support maintaining the status quo including local adoption of the definitions of general and specific provisions. We also agree that guidance from local regulators is required to ensure that general provisions added to Tier 2 capital are restricted to provisions relating to unidentified future losses. We believe that any future changes to the regulatory treatment of accounting provisions should be carefully considered by the Committee and also subject to sufficient consultation with industry. 3. Possible transitional arrangements (p. 4) 3.1 The Committee s primary considerations (p. 4-6) Transitional arrangement Due to the complexity of the ECL models, it is difficult to forecast impacts accurately at this point in time. As such, we believe a transitional arrangement should be allowed to be phased-in over an acceptable period of time. We support a 5-year transitional period and a simple transitional methodology that is easily understood by stakeholders. Date: January 13, 2017 Page 1 of 4

4 Banks stakeholders are particularly focused on the CET1 ratio since the adoption of Basel III. We would suggest that the Committee focus any transitional arrangements on the impact to CET1. Other than dampening the Day 1 impact of IFRS 9 for regulatory capital, the other important rationale for having a transition adjustment is to ensure a level playing field with the US GAAP banks that are only moving to ECL accounting in 2020/21. Disclosure We agree that disclosure is an important element. Care must be taken in the disclosure provisions associated with the transitional arrangements to avoid unintended consequences of such disclosure. While we are supportive of the disclosure of an institution s application of transitional arrangements, given the uncertainty associated with the magnitude of impact of the application of ECL accounting on capital requirements, any additional quantitative disclosure may present challenges or potentially cause undue market misinterpretation. We do not support disclosure of the CET 1 ratio on a fully-loaded basis. A requirement to disclose on a fully-loaded basis may nullify any transitional benefits as investors/analysts would look through the transitional figures. Disclosure requirements should be designed such that the transitional results are presented as the primary focus. Banks reported regulatory capital ratios are those that include the transitional benefit. Readers should not be confused by 2 sets of ratios. We note that Canadian banks currently disclose Basel III all-in ratios but these ratios reflect grandfathering of non-qualifying capital as allowed by OSFI. This has been accepted by investors/analysts and the disclosure of the true all-in (fully loaded) capital ratios is not required. We suggest that the disclosure requirements be limited to Pillar 3 disclosure of the dollar amount of the transitional arrangement and the number of years over which the transitional arrangement is being amortized. This would be consistent with the approach Canadian banks followed for the adoption of IFRS in Q when Canadian GAAP was transitioned. 3.3 Approach 1: Day 1 impact on CET1 capital spread over a specified number of years (p. 6) We support Approach 1. A transitional adjustment calculated on the Day 1 impact amortized over a period of 5 years on a straight line basis is in line with the transitional adjustment from initial IFRS implementation and would be easy for external users to understand. We agree that there would be minimal benefits from a complex transitional impact calculation due to its linkage with provisions, exposures, deferred tax assets, shortfall deductions, etc. We seek clarification on whether the number of years for the transition period will be stated in the final rules as opposed to number of years specified by the Committee. Date: January 13, 2017 Page 2 of 4

5 3.3.1 Calculation of transitional adjustment amount (p. 6-7) We suggest the Committee include text to highlight that an existing shortfall in allowances would not be reflected in the transitional amount. This is currently only determinable by reviewing the example provided. An institution should be given the option of recognizing the full impact on Day 1 versus adopting a transitional approach Mechanics of the transitional arrangement (p. 7) We request further clarification on the example provided for the transitional arrangement under Approach 1. While we support a 5-year phase-in period, if the Committee decides to adopt a 3-year phase-in, the example shows that ¼ of the amount would have been amortized at day 1 (i.e. ¼ of the amount to be included at each reporting date for 3 years) A possible modified version of Approach 1 (materiality threshold) (p. 7-8) We do not support the use of a materiality threshold. Transitional arrangements should be on the full impact of IFRS 9/CECL and not limited to a decrease greater than a "modest" amount since defining a "modest" amount would vary by jurisdiction. Depending on the size of the amount, this can also create an un-level playing field. From an international perspective, it appears unfair to set a threshold based on a bank s capital level relative to its regulatory minimums as this approach is unfair and punitive to highly capitalized banks subject to a higher materiality threshold. Although the change in provisions varies across banks, this is a significant change in accounting policy for all banks and all banks should benefit equally from this transitional adjustment phase-in. We believe that a materiality threshold would also introduce additional complexity to the transitional arrangement calculation and potentially cause capital planning complications. 3.4 Approach 2: CET1 capital adjustment linked to Day 1 proportionate increase in provisions (p. 8) We prefer Approach 1 rather than Approach 2 and believe that Approach 2 should only be considered as a secondary option. Date: January 13, 2017 Page 3 of 4

6 Approach 2 has the advantage of being somewhat dynamic as it does consider the evolution of provisions though time. However, it also assumes that the proportion of total provisions due to ECL accounting remains constant over time which is false. Moreover, it may be complicated to implement. In addition to being overly complex, this approach results in too much volatility in the transitional adjustment and its impact on provisions, exposures, shortfall deductions, deferred tax assets, etc. This approach may also not be easily understandable for investors. Should the Committee decide to adopt Approach 2, however, we request that the Committee clarify if during a subsequent interim period a bank could continue the transitional arrangement if in one period its transition amount is determined to be zero. 3.5 Approach 3: Phased prudential recognition of IFRS 9 Stage 1 and 2 provisions (p. 8-9) We do not believe that Approach 3 should be adopted because it cannot be applied to those reporting under FASB accounting rules (i.e. it will have limited applicability). We also believe that this approach is theoretically weak as it makes the assumption that all Stage 1 and Stage 2 provisions are new. This ignores the current incurred loss provisioning already taking place and reflected in capital. Annex 1 Accounting and regulatory expected loss models (p ) The comparison tables for ECL imply that the lifetime ECL under IFRS 9 and CECL are the same, but they are not. We recommend that the BCBS consider adding a footnote to the tables to explain that CECL lifetime does not equal the IFRS lifetime model. Date: January 13, 2017 Page 4 of 4